Hacker News Reader

Comment by mrfusion | Hacker News Reader

They should consider a kickstarter for funding. Does anyone know if they've considered it?

You need real money to develop aerospace technology. The SABRE engine is in its final stage of development, and according to the articles linked from the main article, just finishing the final stage will cost about 200 million pounds ($330 million dollars). That's about thirty times what the largest project in Kickstarter history has ever raised.

The article pegs the total development cost of the plane at the cost of developing an Airbus plane. That's about $10-15 billion, or more than an order of magnitude more than all of the money that has ever been raised in Kickstarter history. Probably more than all the money that has ever been raised by crowd funding period.

Wow, I had no idea it was that much.

Sorry for the naive question, but any ideas on why it costs so much? Wouldn't putting 100-200 top engineers on it for a few years be enough? What do they need 300 million for?

$300 million to cover "a few years" arguably isn't that much - even basic stuff as the loaded costs of staff is going to be a significant proportion of that.

To maybe put that in perspective, $100 million a year is the budget of one of the smaller F1 teams.

You don't merely need engineers to design the thing. You also have to actually build what you design, custom manufactured out of exotic materials in small batches, and then test those components (often to destruction) before building a slightly-more ambitious prototype. A test or launch also involves a large land area for safety, significant support staff handling hazardous material, and miles of red tape to secure the firing range and airspace.

This build-test-debug loop costs millions and may take months or years, and a single spectacular failure can sink the entire project (e.g. DC-X).

Most engineering is not like software! For example, most of the cost of a bridge is not the civil engineering to design the structure...

> Most engineering is not like software!

Not just that, but original R&D is much more expensive than implementation of proven designs. Not knowing which designs will work, or whether the overall idea will even work at all, greatly increases the number of design/build/test/debug cycles needed to build a product.

Reply on news.ycombinator.com